This is a post with reports and videos on MS SQL Server
In June, we conducted a Microsoft SQL Server mitap — speakers from Yandex.Money, Kaspersky Lab, AwaraIT and SpbDev shared their experience with colleagues. And now we post the video to share.
Under each video - timecode key locations, but we recommend to watch, listen and listen to the whole.
Data-driven Subscriptions in SSRS
Anton Belokrylov, Yandex.Money
Every morning, Yandex.Money managers receive in the mail the financial results of their projects over the past 24 hours. This is a typical scenario of using report subscriptions in SQL Server Reporting Services, which is common in many companies. But we went further: we organized a queue for generating and sending heavy reports and implemented “event-based” generation of database unloadings based on data-driven subscriptions. Anton tells how we did it.
Standard Report Subscriptions
00:15 We use SSRS. Why?
01:50 What does a report work look like?
03:13 How to get a subscription report?
06:30 How to disable sending a blank report?
07:30 Where is the subscription information?
08:40 About variable report parameters?
10:00 How to create multiple reports?
Data driven subscriptions
11:40 How to get a subscription report?
13:20 Example of generated parameters for Data-driven subscriptions
14:30 Benefits of Data-driven Subscriptions
15:15 About the problems
Data driven queued subscriptions
17:50 Event-filled queue model
20:30 What happens if the sources do not match?
23:30 Scheme of work with the queue
26:00 About the benefits
Microsoft PowerBI as an alternative to SQL Server Reporting Services: instead of or together?
Dmitry Vlasov, Awara IT
Microsoft Power BI as an alternative to SQL Server Reporting Services: instead of or together? What is the difference between these two technologies and what is their future? We discuss the features of each, the nuances of licensing, compare performance.
During the recording of this report, something went wrong, but this information content did not suffer. Recommended!
02:26 PowerBI device and usage scenarios
04:15 What if your organization wants PowerBI, but does not want to go to the cloud?
13:32 How is PowerBI useful in different departments - from manager to support?
Important of questions:
16:30 About using the R language in visualization
18:00 How to explain to management that PowerBI should be implemented in a company?
20:25 About AzureStack and data storage in Russia
21:15 About embedding PowerApps in reports
24:30 About personification and distribution of permits within the company
26:40 Is there a good way to use PowerBI on a Mac other than RDP?
Optimizing parallel queries in SQL Server
Sergey Olontsev, Kaspersky Lab
We understand how the parallel execution of queries inside SQL Server occurs. We look, what algorithms and plans of requests are used. We find out how to increase the efficiency of requests. Using examples from real work together we will go from the analysis of the query plan to its acceleration.
01:20 Parallelism in SQL Server
03:35 What is a parallel execution plan for queries in SQL Server
04:30 Concurrency Operators - Distibute, Repartition, Gather
08:00 Two approaches to parallel sorting
10:05 Parallel aggregation - all work at once, or partially, and then connect?
11:10 Demo - how concurrency speeds up sorting
28:00 How does data distribution work in parallel zones?
31:00 Demo - “Parallel execution plan”
34:20 Degree of Parallelism
36:10 What prevents a parallel plan?
37:40 Demo - Some more experiments with parallelism
ETL process refactoring with many levels of nesting
Roman Dimenko, SpbDev
We discuss four steps towards refactoring an ETL process:
- Figure out how the current ETL works;
- Write a new ETL, simple and straightforward;
- Compare the work of the old ETL and the new, correct errors in the business logic;
- Configure continuous integration and test coverage for the new ETL.
01:00 Why a client asked for help. Environment, conditions, requirements.
05:25 About the project. Team, goals.
08:10 How to start refactoring?
10:25 The problems we are facing
13:20 Why do not you need to come up with universal solutions
14:05 How to overcome difficulties?
15:20 How to plan when nothing is known?
17:25 Everything is measurable, if you know the measure
On this concentrate MS SQL dried up. But don't be discouraged - come to events , learn SQL and subscribe to our blog.
')
Source: https://habr.com/ru/post/420087/
All Articles